8 ms·
> Some compression algorithms support checkpointing the compression state so that it can be resumed from that point repeatedly ("dictionary-based compression"
by 1996 5y ago
> Some compression algorithms support checkpointing the compression state so that it can be resumed from that point repeatedly ("dictionary-based compression"
Is it some kind of memoization?
- samus 5y agoAnother toplevel comment claims it is relevant for the use case where you stuff the whole corpus into a single stream. When you want to add new data, you don't want to start over compressing everything. https://news.ycombinator.com/item?id=27441474 https://news.ycombinator.com/item?id=27441474